Accessing multimetallic complexes with a phosphorus(I) zwitterion

Stephanie C. Kosnik , Maxemilian C. Nascimento , Justin F. Binder and Charles L. B. Macdonald *
Department of Chemistry and Biochemistry, The University of Windsor, 401 Sunset Avenue, Windsor, ON N9B 3P4, Canada. E-mail: cmacd@uwindsor.ca

Received 12th October 2017 , Accepted 21st November 2017

First published on 23rd November 2017


We present the synthesis of a zwitterionic triphosphenium molecule, tBu(C5H2)(PPh2)2PI (L), which can act as a single- or multidentate ligand with group 6, 7, 8 and 9 metal carbonyl complexes. Group 6, [M(CO)5L] complexes are formed under photolytic conditions, where the metal is bound at the P(I) center. In the case of Mo(CO)6, the bimetallic complex [M(CO)5LMo(CO)3] is generated, which features bonds to both the phosphorus(I) center and the cyclopentadienyl moiety of the molecule. Interestingly, group 7 and 9 metal carbonyl dimers generate bimetallic complexes in the form [M2(CO)nL], where both metal centers are bound at the phosphorus(I) center. A detailed analysis of these molecules is provided, including X-ray diffraction, multinuclear NMR, infrared spectroscopy and computational investigations.


Introduction

The synthesis of multimetallic complexes has been an important topic in organometallic chemistry, both in the areas of catalysis,1,2 and in ligand design.3,4 Typically, bulky, pincer-type ligands or macrocycles are used for these applications, often featuring various main group donors or acceptors (B, C, N, P, S, O).3,5–8 An alternative approach to generate multimetallic species is to employ ligands containing low valent main group centers, which may bind multiple metal centers at one atom due to the increased electron density and low coordination number about the low valent main group center. The low valent ligand approach may reduce the steric bulk in the system, and perhaps increase the activity of the transition metal center, or allow two metals in close proximity for cooperative catalysis.2,9 There are several classes of low valent or low coordinate phosphorus centers that have been used to coordinate transition metals, including phosphides,10–14 phospholides,15,16 phosphinines,17–24 phosphaalkenes,25–29 isophosphindoliums30–33 and even phosphanes,34 among others. One of the most extensively investigated classes of neutral ligands featuring low valent phosphorus centers are the phosphinidenes, the phosphorus analogue of carbenes or nitrenes. These ligands have been comprehensively studied by many, and have been shown to create numerous stable mono- and bimetallic complexes with a variety of transition metals35–44 (Fig. 1A and B).
image file: c7dt03844e-f1.tif
Fig. 1 Molecules containing a low valent phosphorus center which can be capable of coordinating transition metals.

Of particular relevance to the work presented herein are the bimetallic complexes with a low valent main group complexes reported by Frenking45 and Alcarazo using carbodiphosphorane (R3P-C-PR3) ligands. They demonstrated that these formally carbon(0) compounds can accommodate up to two gold(I) fragments bound by a single carbon(0) center46 (Fig. 1C). Similar reactivity was demonstrated by Ragogna et al. using a zwitterionic triphosphenium47 ligand that contains an isolobal phosphorus(I) center (which was also able to coordinate up to two gold atoms on the PI center (Fig. 1D1 & D2)).48 The use of neutral zwitterionic analogues of triphosphenium cations for coordination chemistry is necessary because, in spite of the “phosphanide” fragment present in a triphosphenium cation, (e.g.Fig. 1E) these cations appear to be unsuitable for the generation of persistent metal complexes.49,50

We have also been interested in the coordination chemistry of zwitterionic triphosphenium species, and in their potential use as multidentate donors, both in the form of macrocycles51 and molecules which feature multiple coordination sites.52 In the former case, we observed that the coordination of late transition metals occurs via both low coordinate phosphorus centers in a κ-P,P′-bidentate type fashion, while in the latter case we observed that the reactivity of the molecule is localized – surprisingly – on the pendant phosphine group rather than at the low coordinate phosphorus center or the cyclopentadienyl fragment (Fig. 1F). We posited that we could better engage the cyclopentadienyl fragment within the backbone by replacing the phosphine group with a coordinately “inert” alkyl group instead. This would allow us to achieve bimetallic complexes through either the PI center alone or via both the Cp moiety and PI center together. The successful results of the improved ligand design are presented in this work.

Results & discussion

Synthesis and characterization of [tBuC5H2(Ph2P)2PI]

We previously reported the synthesis of a zwitterionic triphosphenium molecule that featured three potential coordination sites. In our initial report, we had noted that the reactivity of the molecule occurred preferentially on the terminal phosphine group attached to the cyclopentadienyl fragment rather than the phosphorus(I) centre, and this result was rationalized on the basis of electrostatic potential analysis derived from density functional theory investigations.52 With this insight, we set out to modify the ligand so as to enhance the reactivity of the phosphorus(I) moiety; one obvious modification was to eliminate the terminal phosphine site and to utilize instead an alkyl-substituted cyclopentadienyl backbone – we selected a tert-butyl substituted analogue – that would still allow for the convenient preparation of the 1,2-diphosphine ligand. We were able to generate the lithium 1,2-bis(diphenylphosphino) tert-butylcyclopentadienide ligand using a literature method53 and we found that the reaction of this ligand with our PI transfer agent, [PIdppe][Br], generates the desired product, 1, in good yields (Scheme 1). The removal of the two by-products (dppe and LiBr) was accomplished using a two-step process. The dppe was removed by Soxhlet extraction of the crude mixture in hexanes overnight, and then the remaining solid was dissolved in dichloromethane and filtered through a fritted flask to remove the LiBr. The pure material has distinctive 31P NMR shifts consistent with those reported for triphosphenium species: a very shielded triplet (−174 ppm) and corresponding doublet, (34 ppm) with large 1JPP couplings (418 Hz). Single crystals were grown by the slow evaporation of a dichloromethane solution, and the molecular structure is depicted in Fig. 2. The air-sensitive molecule crystallizes in the space group P[1 with combining macron] with one molecule present in the asymmetric unit. The P–P bond lengths of the triphosphenium fragment of 2.1260(9) and 2.1303(9) Å are within the range of distances reported for cationic cyclic triphosphenium cations (2.11–2.13 Å) – and are short in comparison to those of other zwitterionic triphospheniums that have been reported previously (2.1328(9) to 2.1467(12) Å).48,52 The P–PI–P angle of 89.76(4)° falls in between the ranges of the typical angles observed for zwitterionic triphosphenium fragments, (90.39(4)–95.70(3)°) and triphosphenium cations (86–88°).
image file: c7dt03844e-s1.tif
Scheme 1 The reaction of [Li][tBuC5H2(Ph2P)2] with [PIdppe][Br] to generate 1.

image file: c7dt03844e-f2.tif
Fig. 2 The thermal ellipsoid plot of 1, ellipsoids depicted at 70% probability, hydrogen atoms removed for clarity. Selected bond lengths and angles: P3–C2: 1.7342(19), P1–P2: 2.1303(9), P2–P3: 2.1260(9), P1–C1: 1.7507(19), P3–C2: 1.7342(19), C1–C5: 1.398(3), C1–C2: 1.417(2), C2–C3: 1.401(3), C3–C4: 1.401(3), C4–C5: 14.05(2) Å. P1–P2–P3: 89.76(4)°.

Reactions of group 6 metal carbonyls with 1 [tBuC5H2(Ph2P)2PI]

In light of some of the previous coordination chemistry of zwitterionic compounds containing PI centres – and in light of the potential for compound 1 (indicated as L in the formulas of the complexes) to also support piano stool complexes – we began our studies with early zero-valent transition metal carbonyl complexes. The addition of excess M(CO)6 (M = Cr, Mo, W) to 1 in THF under UV light for 1–3 hours generates the corresponding coordination complexes 2–3 (Scheme 2). Upon irradiation, the light yellow coloured solutions gradually turn dark orange or dark gold in colour, and the progress of the reaction can be monitored by 31P NMR. All the complexes have dramatic deshielding of the triplet shift corresponding to the PI centre upon complexation (−174 ppm to −51 ppm, −76 ppm, and −91 ppm for M = Cr, Mo, and W respectively), in addition to a significant decrease in the coupling constants by 50–70 Hz (Fig. 3). When the reaction has gone to completion, the reaction solvent is removed under reduced pressure and the excess M(CO)6 can be removed via sublimation. The pure product was isolated by extraction into Et2O, and the resultant solution was left to evaporate yielding large dark gold-coloured crystals. Single crystal X-ray diffraction experiments reveal that the anticipated complexes [LCr(CO)5] 2 and [LW(CO)5] 3 are isostructural (Fig. 4). Both molecules crystallize in the space group P[1 with combining macron] with one molecule in the asymmetric unit. As anticipated, upon coordination of the PI center, there is a slight lengthening of the P–P bond within the triphosphenium fragment, from 2.1260(9)–2.1303(9) to 2.1711(5)–2.1759(5) Å for 2 and 2.1656(13)–2.1696 (13) Å for 3. The phosphorus–metal distances are 2.4420(5) Å, and 2.5799(7) Å for Cr and W respectively, and are at the longer end of the ranges of bond lengths of related dicoordinate phosphorus atoms bound to Cr(CO)5 or W(CO)5 reported in the CSD, (2.302–2.523 Å for Cr and 2.454–2.675 Å for W).54 In each complex, the geometry about the coordinated phosphorus center is trigonal pyramidal, which is consistent with the presence of an additional pair of non-bonding electrons at the PI centre; the sum of the angles around the PI centre is 334.40(3)° and 333.38(8)° for the chromium and tungsten complexes, respectively.
image file: c7dt03844e-s2.tif
Scheme 2 The reaction of 1 with Cr(CO)6 and W(CO)6 under UV light.

image file: c7dt03844e-f3.tif
Fig. 3 31P {1H} stack plot of complexes 1, 2, 3, 5 and 7.

image file: c7dt03844e-f4.tif
Fig. 4 The thermal ellipsoid plot of 2 (left) and 3 (right), ellipsoids depicted at 50% probability, hydrogen atoms removed for clarity. Selected bond lengths and angles for 2: P1–P3: 2.1759(5), P2–P3: 2.1711(5), P3–Cr: 2.4420(5) Å. P1–P3–P2: 91.081(19)°. Selected bond lengths and angles for 3: P1–P3: 2.1696(13), P2–P3: 2.1656(13), P3–W: 2.5601(10) Å. P1–P3–P2: 91.36(5)°.

Compound 2 exhibits an axial CO bond length of 1.146(2) Å which is similar in length to the average length of the equatorial CO ligands: 1.139(2) Å. The same is true of 3, which exhibits CO bond lengths of 1.138(6) Å (axial) and 1.132(6) Å (equatorial). The M–CO bonds range from 1.823(19)–1.9046(17) Å (Cr) and 1.989(5)–2.038(5) Å (W), with the axial M–CO bond being shorter than the equatorial bonds in all cases. The FT-IR spectra of both compounds (Fig. 7) feature multiple carbonyl stretches between 1895 cm−1 to 2055 cm−1 (2) and 1913 cm−1 to 2065 cm−1 (3). The appearance of more peaks than what might be expected for a C4v symmetric complex is due to the non-degeneracy of the usual E modes, which are split significantly due to the asymmetry introduced by the ligand; this also permits the appearance of the normally forbidden B1 mode, which gains some intensity. To confirm the assignment of the peaks, frequency calculations were performed on a model of 5 optimized at the PBE1PBE/TZVP level of theory with an effective core potential basis set applied for the metal atom (ESI Fig. 10 and 15).

The FT-IR spectrum of the material isolated from the reaction of 1 with excess Mo(CO)6 under photolytic conditions (4) shows five similar carbonyl stretches (1896 cm−1 to 2067 cm−1) to that observed in the spectra of 2 and 3, however, there were three additional weaker stretches at 1804 cm−1, 1814 cm−1, and 1831 cm−1 present in the spectrum for 4 which were not observed in the spectra for 2 and 3 (Fig. 7). Additionally, the calculated frequencies for the complex [LMo(CO)5] do not account for the three additional stretches present in the spectrum (ESI Fig. 11).

The single crystals obtained from slow evaporation of the Mo(CO)6 reaction revealed an unexpected result: a bimetallic molecule (4), featuring a Mo(CO)5 fragment coordinated at the low valent phosphorus centre and an additional Mo(CO)3 fragment ligated through the cyclopentadienyl backbone in a piano-stool fashion (Fig. 5). Despite using identical reaction conditions to those that produced 2 and 3, only the bimetallic species is observed based on the NMR and FT-IR spectra obtained, and from unit cell analysis of the resulting single crystals. Complex 4 crystallizes in the space group Pbca, with one molecule in the asymmetric unit, accompanied by a solvent molecule (Et2O). The P–P bond lengths of the triphosphenium fragment, display a similar same lengthening upon complexation of the metal center, in addition to an increase in the P–PI–P angle to 91.65(3)°. Ligation of the additional metal centre at the cyclopentadienyl fragment has similar effects; the C–C bonds within the five-membered ring increase in length from 1.398(3)–1.405(3) Å in 1 to 1.441(3)–1.420(3) Å in 4. The average M–CCp bond length within the piano stool is 2.398(2) Å, which is typical of lengths reported for other Mo(CO)3 piano stool complexes reported in the CSD,54 (2.308–2.425 Å). The Mo–CO distances are also typical and range from 1.940(3) to 1.951(3) Å, with the axial Mo–CO bond being shorter than the equatorial bonds in the Mo(CO)5 fragment, as anticipated. The three weaker frequency vibrations observed in the IR spectrum obtained for 4 (vide supra) are attributable to the E stretching modes of the Mo(CO)3 fragment which is bound by the Cp-moiety (Fig. 7) and are corroborated by calculations (ESI, Fig. 12).


image file: c7dt03844e-f5.tif
Fig. 5 The thermal ellipsoid plot of 4, ellipsoids depicted at 50% probability, hydrogen atoms and diethyl ether molecule removed for clarity. Selected bond lengths and distances: P1–P3: 2.1813(8), P2–P3: 2.1743(9), P3–Mo1: 2.5799(7), Mo2-centroid 2.045 Å. P1–P3–P2: 91.65(3)°.

In an effort to isolate the monometallic molybdenum carbonyl species, i.e. the molecule that contains only Mo(CO)5 coordinated that the PI center, we first reacted the ligand, 1 with Mo(CO)6 in an equimolar (1[thin space (1/6-em)]:[thin space (1/6-em)]1) ratio, under photolytic conditions. The reaction proceeds quickly (within 1 h), and the triplet in the 31P NMR corresponding to the PI centre in 1 shifts from −174 ppm, to −76 ppm, however some starting material remains, even after several hours in the UV reactor. Upon isolation of this reaction product, we confirmed via IR spectroscopy and single crystal diffraction that it was indeed 4, rather than the monometallic complex that we were targeting, so we attempted an alternative route to generate the molybdenum analogue of 2 and 3. Thus trimethylamine-N-oxide was added to Mo(CO)6 in THF in the presence of 1 (Scheme 3). The reaction mixture became gold in colour and small bubbles were observed indicative of the loss of CO2. 31P NMR spectroscopy confirmed that all the starting material (1) was consumed in the reaction, and the new doublet and triplet signals in the spectrum were indistinguishable from those observed for 4. The THF was removed under reduced pressure, and product extracted with Et2O. FT-IR spectroscopic analysis of the resulting yellow solid provided a spectrum in which the three unique IR stretches corresponding to the Mo(CO)3 fragment were absent; only the five stretches attributable to the Mo(CO)5 fragment were present in the carbonyl region of this spectrum (1895–2067 cm−1). Single crystals obtained from this reaction confirmed the molecular structure of [L(Mo(CO)5] 5 (Fig. 6), which is isostructural to 2 and 3. The molecule, which crystallizes in the space group P[1 with combining macron], shows similar lengthening of the P–P bonds and P–PI–P angle within the triphosphenium fragment, observed in the other metal complexes described above (see Table 1).


image file: c7dt03844e-s3.tif
Scheme 3 Synthesis of molybdenum carbonyl complexes from 1.

image file: c7dt03844e-f6.tif
Fig. 6 The thermal ellipsoid plot of 5, ellipsoids depicted at 50% probability, hydrogen atoms removed for clarity. Selected bond lengths and distances: P1–P3: 2.1635(4), P2–P3: 2.1677(4), P3–Mo1: 2.5734(4) Å. P1–P3–P2: 91.344(7)°.
Table 1 Summary of bond lengths, angles, and 31P NMR shifts for 1 and its complexes its group 6 metal carbonyls
Parameter 1 (L) 2 [L(Cr(CO)5)] 3 [L(W(CO)5)] 4 [L(Mo2(CO)8)] 5 [L(Mo(CO)5)]
P–PI (Å) 2.1303(9) 2.1759(5) 2.1696(13) 2.1813(8) 2.1635(4)
2.1260(9) 2.1711(5) 2.1656(13) 2.1743(9) 2.1677(4)
P–PI–P (°) 89.76(4) 91.081(19) 91.37(5) 9.165(3) 91.344(17)
PI–M 2.4420(5) 2.5601(10) 2.5799(7) 2.5734(4)
∑∠PI (°) 334.40(3) 333.38(8) 325.05(5) 334.69(2)
δ 31P (ppm) −174 (t), 34 (d) −51 (t), 28.8 (d) −91 (t), 25.9 (d) −77 (t), 28.2 (d) −76 (t), 28.8 (d)
1 J p–p (Hz) 418 376 356 356 356
M–COeq (Å) 1.8996(18) 2.038(6) 2.0455(3) 2.4555(17)
M–COax (Å) 1.8523(19) 1.989(5) 1.987(3) 1.9849(17)


The IR stretching frequencies observed for molecules 2–5 (Fig. 7) places the donor ability of this ligand between that of amines and phosphines.55–60


image file: c7dt03844e-f7.tif
Fig. 7 FT-IR spectrum of 1, 2, 3, 4, and the dashed lines indicate the symmetry modes that are assigned to each stretching frequency. The * label indicates the stretches that correspond to the piano-stool moiety in 4, which are E stretching modes.

Orbital depictions of optimized LM(CO)5 models show significant contributions from the Cp fragment on the three highest energy MOs (Fig. 8 and ESI). The relative energies of these orbitals are also very similar which suggests that the difference in observed reactivity between the Mo and two other carbonyl complexes is more likely caused by the increased photoactivity of Mo(CO)6 and the larger kinetic lability of the CO ligands on Mo.


image file: c7dt03844e-f8.tif
Fig. 8 Frontier orbitals of geometry optimized structure of 4.

Thus far, our efforts towards the generation of bimetallic W2(CO)8 and Cr2(CO)8 analogues to 4 have been unsuccessful. Even under harsher conditions – UV radiation for 48 h, and large excesses of metal carbonyl (10 equivalents) – there is no evidence for the formation of any bimetallic complexes or ligand displacement.

Reactions group 7, 8, and 9 metal carbonyls with [tBuC5H2(Ph2P)2PI]

We were curious about how the reactivity of this multidentate ligand might change to accommodate different zero-valent metal carbonyls moving across the first row of the periodic table. The equimolar reaction between manganese, Mn2(CO)10, did not proceed after stirring under standard conditions overnight. However, there was a significant colour change from pale yellow to bright red upon stirring under UV radiation for 3 hours. The reaction progresses to completion after stirring under photolytic conditions for 6 hours, and the THF was removed from the solution. The 31P NMR spectrum of the resulting red solution showed a significant change from that of the proligand: the triplet corresponding to the PI center shifts from −178 ppm to +183 ppm, along with a significant decrease in 1JPP coupling (418 Hz to 283 Hz). Broadening of the triplet corresponding to the PI center, along with a dramatic change of its shift by more than 200 ppm, and a decrease in 1Jpp coupling was our first indication that both manganese metal centers were bound by the PI center. The product was extracted with Et2O and left to evaporate slowly which lead to the deposition of large red crystals. The bimetallic complex [L(Mn2(CO)8)] 6 crystallizes in the space group P[1 with combining macron] with one molecule in the asymmetric unit (Fig. 9). The PI center of 1 acts as a μ-ligand across two Mn atoms, utilizing both available pairs of electrons on the low valent phosphorus center. Both the Mn–Mn bond length, (2.803(2) Å) and the P–Mn bond lengths (2.2604(5) and 2.2720(5) Å) are within the ranges of distances (Mn–Mn: 2.706–3.070 Å and P–Mn: 2.176–2.380 Å) for complexes featuring Mn2(μ-PR2) fragments reported in the CSD. As anticipated, there is a substantial lengthening of the P–P bonds within the triphosphenium fragment upon complexation of the metal: from 2.1260(9)–2.1303(9) in 1 to 2.2322(2)–2.2410(6) – these distances are considerably longer than any of the other metal complexes reported in this work. The FT-IR spectrum of 6 has five frequencies corresponding to carbonyl stretches ranging from 1911 to 2050 cm−1.
image file: c7dt03844e-f9.tif
Fig. 9 The thermal ellipsoid plot of 6, ellipsoids depicted at 50% probability, hydrogen atoms and diethyl ether molecule removed for clarity. Selected bond lengths and angles: P1–P3: 2.2322(6), P2–P3: 2.2410(6), P3–Mn1: 2.2720(5), P3–Mn2: 2.2604(5) Å. P1–P3–P2: 88.88(2)°.

In contrast to the reaction to generate 6, the addition of Fe2(CO)9 to 1 and the subsequent reaction under photolytic conditions yielded many species observable by 31P NMR in the form of an intractable mixture. In contrast, the reaction between the less reactive iron(0) source, Fe(CO)5, lead to no reaction under both standard, thermolytic and photolytic conditions. Even the use of the trimethylamine-N-oxide protocol described earlier to generate 5 generated three different phosphorus-containing products based on 31P NMR spectroscopic investigations (ESI Fig. 7); all of these products have similar solubilities which rendered their separation and isolation difficult. Finally, we found that the reaction of 1 and Fe2(CO)9 under ambient conditions did generate only one product based 31P NMR experiments, but the reaction proceeded slowly. After 4 days stirring in THF, all of 1 was consumed and the distinctive triplet from the proligand (−174 ppm) shifts significantly downfield to −16 ppm. It should be noted that there is an additional singlet present in the spectrum at 77 ppm, which we suspect is tBuCp(PPh2)2Fe(CO)4i.e. the iron diphosphine complex derived from the formal extrusion of the phosphorus(I) fragment from 1. The formation of a similar complex was observed in the reaction of an analogous arsenic species with Co2(CO)8, as reported by Ragogna et al. (the 31P NMR shift for their complex resided at 43 ppm).61 Nevertheless, work up of the reaction by centrifugation and extraction with pentane yields the coordination complex [LFe(CO)4] 7. Yellow single crystals were obtained from the slow evaporation of hexanes, and X-ray diffraction experiments reveal the molecular structure of 7, (Fig. 10) which crystalizes in the space group P21/c with one molecule in the asymmetric unit.


image file: c7dt03844e-f10.tif
Fig. 10 The thermal ellipsoid plot of 7, ellipsoids depicted at 50% probability, hydrogen atoms and diethyl ether molecule removed for clarity. Selected bond lengths and angles: P1–P3: 2.1982(6), P2–P3: 2.1871(6), P3–Fe1: 2.2869(5) Å. P1–P3–P2: 90.17(2)°.

As with the group 6 complexes described earlier in this work, the bond lengths and angles of complex 7 exhibit similar features: there is a slight shortening of the P–P bond lengths (2.1982(6)–2.1871(6) Å), a small increase in the P–P–P bond angle (90.17(2)°), and the complex features a P–Fe bond length of 2.2869(5) Å that is in good agreement with those reported for R2P–Fe complexes in the CSD (2.202–2.415 Å). A summary of these parameters can be found in Table 2. Complex 7 has a FT-IR spectrum that is consistent with the mononuclear complex and features frequencies in the carbonyl region between 1930 and 2033 cm−1.

Table 2 Summary of bond lengths, angles, and 31P NMR shifts for 1 and its complexes with group 7, 8, and 9 metal carbonyls
Parameter 1 (L) 6 [L(Mn2(CO)8)] 7 [L(Fe(CO)4)] 8 [L(Co2(CO)8)]
P–PI (Å) 2.1303(9) 2.2322(6) 2.1982(6) 2.214(2)
2.1260(9) 2.2410(6) 2.1871(6) 2.2166(19)
P–PI–P (°) 89.76(4) 88.88(2) 90.17(2) 91.37(5)
PI–M 2.2604(5), 2.2720(5) 2.2869(5) 2.1306(19), 2.1413(16)
∑∠PI (°) 322.18(34)
δ 31P (ppm) −174 (t), 34 (d) 183.7 (t), 19 (d) −16 (t), 22 (d) 177 (t), 6 (d)
1 J p–p (Hz) 418 283 374 242
M–COeq (Å) 1.818(2) 1.795(2) 1.795(7)
M–COax (Å) 1.831(2) 1.784(2) 1.775(6)


In contrast to the iron carbonyl reactions, the 1[thin space (1/6-em)]:[thin space (1/6-em)]1 stoichiometric reaction between 1 and Co2(CO)8 proceeds immediately, yielding a dark red solution in dichloromethane. The 31P spectrum of this reaction after stirring for 5 min shows complete consumption of the proligand and a new set of broadened peaks at +178 ppm (triplet), and 19 ppm (doublet). Similarities in the 31P NMR spectrum between this reaction and the spectrum corresponding to 6 suggested an analogous dinuclear complex [LCo2(CO)6] 8, (Scheme 4); these observations are also consistent with those of Ragogna et al. for their zwitterionic systems.61 The FT-IR spectrum of 8 shows five distinct frequencies that correspond to the carbonyl ligands on each of the cobalt centres. The resulting solution was centrifuged to remove a dark brown precipitate and the reaction solvent evaporated, depositing dark red crystals.


image file: c7dt03844e-s4.tif
Scheme 4 Synthesis of the bimetallic manganese (6) and cobalt (8) complexes from 1.

X-ray diffraction experiments performed on these single crystals confirmed the synthesis of the bimetallic cobalt carbonyl complex 8, in which both cobalt centres are bound by the single phosphorus(I) center in a manner that is analogous to 6 (Fig. 11).


image file: c7dt03844e-f11.tif
Fig. 11 The thermal ellipsoid plot of 8, ellipsoids depicted at 50% probability, hydrogen atoms and diethyl ether molecule removed for clarity. Selected bond lengths and angles: P1–P3: 2.214(2), P2–P3: 2.2166(19), P3–Co1: 2.2720(5), P3–Co2: 2.2604(5) Å. P1–P3–P2: 91.14(7)°.

In light of the structural similarities between 6 and 8 it is unsurprising that the metrical parameters within the ligand are similar. A significant lengthening of the P–P bond lengths within the triphosphenium fragment upon complexation (to 2.214(2)–2.2166(19) Å) is observed. This longer distance is consistent with the decreased magnitude in 1JPP coupling (from 418 to 242 Hz); together, these results suggest significant donation of electrons from the phosphorus(I) center to the metal centers and decreased backbonding interaction within the triphosphenium core. The Co–Co bond length is 2.6675(13) Å which is at the longer end of the range of reported distances (2.412–2.716 Å) of similar PCo2 complexes. The FT-IR spectrum of 8 is similar to that of 6 and contains five frequencies corresponding to the C–O stretches from 1932 to 2032 cm−1.

Conclusion

In this work, we report a neutral triphosphenium analogue for use as a multifunctional donor to zero-valent transition metals. Photolytic reactions with group 6 metal carbonyls as well as Fe2(CO)9, result in the formation of monometallic complexes in which the low valent phosphorus(I) center within the triphosphenium fragment acts as a donor to the metal carbonyl moiety. For Mo(CO)6, under photolytic conditions, a bimetallic complex is formed selectively, as the ligand is able to bind two molybdenum fragments through both the PI center and the cyclopentadienyl moiety within the ligand backbone. Similarly, the reaction of the zwitterionic triphosphenium with Co2(CO)8 and Mn2(CO)10 generate bimetallic complexes, however in this case both metal centres are bound through the phosphorus(I) center. Experiments to probe the reactivity of this molecule with other classes of transition metal complexes are currently underway.

Experimental

General procedures

All manipulations were carried out using standard inert atmosphere techniques. All chemicals and reagents were purchased from Sigma-Aldrich except for Fe2(CO)9 which was purchased from Strem Chemicals. The group 6 metal carbonyl complexes were sublimed prior to use, and all other materials were used as received, without further purification. Deuterated solvents were dried over activated 3 Å sieves. All other solvents were dried over a series of Grubbs’-type columns and degassed prior to use. The reagents, [PIdppe][Br],62 and [Li][tBuC5H2(Ph2P)2]53 were synthesized according to literature procedures. NMR spectra were recorded at room temperature on a Bruker Avance III 500 MHz or Bruker Avance Ultrashield 300 MHz spectrometer. Chemical shifts are reported in ppm relative to internal standards for 1H and 13C (for the given deuterated solvent) and external standard for 31P (85% H3PO4 = 0 ppm). Elemental Analysis performed using a PerkinElmer 2400 combustion CHN analyser. FT-IR was performed on a Bruker ALPHA FT-IR spectrometer using a platinum ATP sampling module; stretching frequencies are reported in cm−1. Photolysis reactions were performed in a Luzchem Photoreactor (Model: LZC-ICH2) using UVA lamps under conditions listed in the Experimental section. High-resolution electrospray-ionization mass spectrometry was performed at the McMaster Regional Centre for Mass Spectrometry.

Crystallographic details

Crystals for investigation were covered in Nujol®, mounted into a goniometer head, and then rapidly cooled under a stream of cold N2 of the low-temperature apparatus (Oxford Cryostream) attached to the diffractometer. The data were then collected using the APEXIII software suite56 on a Bruker Photon 100 CMOS diffractometer using a graphite monochromator with Mo-Kα radiation (λ = 0.71073 Å). For each sample, data were collected at low temperature. APEX-III software was used for data collection and reduction and SADABS57 was used for absorption corrections (multi-scan; semi-empirical from equivalents). XPREP was used to determine the space group and the structures were solved and refined using the SHELX58 software suite as implemented in the WinGX59 program suites. Validation of the structures was conducted using PLATON.60 The diethyl ether molecule in 6 was modelled using the DSR method63 implemented in ShelXle.64 A full list of the relevant crystallographic details for all molecules reported in this work can be found in Tables 3 and 4.
Table 3 Summary of crystallographic data for compounds 1–4
Compound 1 2 3 4
R 1 = ∑(|Fo| − |Fc|)/∑Fo, wR2 = [∑(w(Fo2Fc2)2)/∑(wFo4)], GOF = [∑(w(Fo2Fc2)2)/(no. of reflns. − no. of params.)]1/2.
CCDC ID 1578538 1578540 1578541 1578542
Empirical formula C33H31P3 C38H31P3O5Cr C38H31P3O5W C45H41P3O9Mo2
Formula weight 520.49 712.54 844.39 1010.57
Temperature/K 170(2) 170(2) 170(2) 170(2)
Crystal system Triclinic Triclinic Triclinic Orthorhombic
Space group P[1 with combining macron] P[1 with combining macron] P[1 with combining macron] Pbca
a 9.845(2) 11.2512(8) 11.3304(11) 12.0912(4)
b 11.647(2) 11.7421(7) 11.7799(11) 20.0536(6)
c 13.057(3) 15.0867(10) 15.1089(14) 37.1612(11)
α 96.65(3) 91.279(3) 91.631(3) 90
β 106.37(3) 101.068(3) 100.807(3) 90
γ 101.39(3) 114.470(2) 114.414(3) 90
Volume/Å3 1384.5(5) 1768.8(2) 1791.0(3) 9010.6(5)
Z 2 2 2 8
ρ calc g cm−3 1.248 1.338 1.566 1.490
μ/mm−1 0.253 0.501 3.4 0.716
F(000) 548 736 836.0 4096
Crystal size/mm3 0.2 × 0.15 × 0.12 0.43 × 0.33 × 0.29 0.35 × 0.29 × 0.03 0.20 × 0.18 × 0.024
2Θ range for data collection/° 3.086 to 27.495 3.055 to 27.569 6.096 to 58.448 2.858 to 26.395
Index ranges −12 ≤ h ≤ 12, −15 ≤ k ≤ 15, 0 ≤ l ≤ 16 −14 ≤ h ≤ 14, −15 ≤ k ≤ 15, −19 ≤ l ≤ 19 −15 ≤ h ≤ 15, −16 ≤ k ≤ 16, −20 ≤ l ≤ 20 −15 ≤ h ≤ 14, −23 ≤ k ≤ 24, −36 ≤ l ≤ 45
Reflections collected 11[thin space (1/6-em)]922 78[thin space (1/6-em)]418 89[thin space (1/6-em)]946 88[thin space (1/6-em)]417
Independent reflections 11[thin space (1/6-em)]922 Twinned 8157 [Rint = 0.0405] 9717 [Rint = 0.0605] 9148 [Rint = 0.0354]
Data/restraints/param 11[thin space (1/6-em)]922/0/328 8157/0/427 9717/0/427 9148/11/547
Goodness-of-fit on F2 1.054 1.031 1.134 1.059
Final R indexes [I ≥ 2σ(I)] R 1 = 0.0363 R 1 = 0.0299 R 1 = 0.0371 R 1 = 0.0318
wR2 = 0.0931 wR2 = 0.0727 wR2 = 0.0898 wR2 = 0.0687
Final R indexes [all data] R 1 = 0.0406 R 1 = 0.0411 R 1 = 0.0478 R 1 = 0.0449
wR2 = 0.0975 wR2 = 0.0786 wR2 = 0.0938 wR2 = 0.0725
Largest diff. peak/hole/e Å−3 0.365/−0.356 0.345/−0.378 4.11/−1.37 0.909/−0.576
Refinement method Full-matrix least-squares on F2
Data completeness 0.998 0.995 0.998 0.991


Table 4 Summary of crystallographic data for compounds 5–8
Compound 5 6 7 8
R 1 = ∑(|Fo| − |Fc|)/∑Fo, wR2 = [∑(w(Fo2Fc2)2)/∑(wFo4)], GOF = [∑(w(Fo2Fc2)2)/(no. of reflns. − no. of params.)]1/2.
CCDC ID 1578543 1578544 1578545 1578546
Empirical formula C38H31P3O5Mo C43H36P3O8Mn2 C37H31P3O4Fe C39H31P3O6Co2
Formula weight 756.48 891.51 688.38 806.41
Temperature/K 170(2) 170(2) 170(2) 170(2)
Crystal system Triclinic Triclinic Triclinic Triclinic
Space group P[1 with combining macron] P[1 with combining macron] P21/c P[1 with combining macron]
a 11.3621(5) 11.1044(6) 16.0868(6) 10.6793(4)
b 11.7938(6) 11.8694(7) 12.0933(4) 11.8419(5)
c 15.1337(8) 16.9388(10) 18.1738(7) 14.9909(7)
α 91.699(2) 105.480(2) 90 87.867(3)
β 100.767(2) 93.283(2) 99.7590(10) 84.483(3)
γ 114.322(2) 99.354(2) 90 78.487(3)
Volume/Å3 1802.51(5) 2111.1(2) 3484.4(2) 1848.73(14)
Z 2 2 4 2
ρ calc, g cm−3 1.394 1.403 1.312 1.499
μ/mm−1 0.538 0.764 0.608 8.632
F(000) 772 914 1424 824
Crystal size/mm3 0.36 × 0.28 × 0.076 0.51 × 0.27 × 0.19 0.28 × 0.15 × 0.04 0.14 × 0.05 × 0.02
2Θ range for data collection/° 5.52 to 61.154 2.799 to 32.647 2.795 to 30.555 2.962 to 65.381
Index ranges −15 ≤ h ≤ 16, −16 ≤ k ≤ 16, −21 ≤ l ≤ 21 −16 ≤ h ≤ 16, −17 ≤ k ≤ 17, −25 ≤ l ≤ 25 −22 ≤ h ≤ 22, −17 ≤ k ≤ 17, −25 ≤ l ≤ 25 −11 ≤ h ≤ 12, −13 ≤ k ≤ 13, −17 ≤ l ≤ 17
Reflections collected 106[thin space (1/6-em)]804 129[thin space (1/6-em)]291 118[thin space (1/6-em)]875 51[thin space (1/6-em)]126
Independent reflections 11[thin space (1/6-em)]029 [Rint = 0.0447] 15[thin space (1/6-em)]412 [Rint = 0.0447] 10[thin space (1/6-em)]530 [Rint = 0.0859] 6131 [Rint = 0.1489]
Data/restraints/param 11[thin space (1/6-em)]029/0/427 15[thin space (1/6-em)]412/48/537 10[thin space (1/6-em)]530/0/409 6131/0/454
Goodness-of-fit on F2 1.050 1.095 1.038 1.172
Final R indexes [I ≥ 2σ(I)] R 1 = 0.0266 R 1 = 0.0445 R 1 = 0.0440 R 1 = 0.0713
wR2 = 0.0669 wR2 = 0.0997 wR2 = 0.0822 wR2 = 0.1102
Final R indexes [all data] R 1 = 0.0337 R 1 = 0.0807 R 1 = 0.0877 R 1 = 0.1194
wR2 = 0.0711 wR2 = 0.1232 wR2 = 0.0939 wR2 = 0.1235
Largest diff. peak/hole/e Å−3 0.47/−0.79 1.197/−0.655 0.568/−0.362 0.546/−0.551
Refinement method Full-matrix least-squares on F2
Data completeness 0.997 0.997 0.986 0.965


Computational details

Calculations were performed with the Gaussian 09 suite of programs65 using Compute Canada's Shared Hierarchical Academic Research Computing Network (SHARCNET). Model complexes were fully optimized with no symmetry constraints using the PBE1PBE density functional theory (DFT) method66–68 in conjunction with the TZVP basis sets for all atoms.69,70 The default Stuttgart–Dresden (SDD) quasi-relativistic effective core potentials were used for transition element atoms.71,72 Geometry optimizations were started using models in which the relevant non-hydrogen atoms were placed in positions found experimentally using X-ray crystallography and the hydrogen atoms were placed in geometrically appropriate positions using Gaussview.73 Frequency calculations were also performed at the same level of theory in order to confirm that the optimized structures were minima on the potential energy hypersurface and to determine thermochemical and vibrational information. Natural bond orbital (NBO) analyses74 to determine orbital contributions, Wiberg bond indicies and orbital energies were obtained using the routine included in the Gaussian distributions.75 Visualizations of the Kohn–Sham orbitals and optimized geometries were made using Visual Molecular Dynamics (VMD).76 Summaries of the calculated results, including Cartesian coordinates are presented in the ESI.

Synthesis of tBuCp(PPh2)2PI (1)

To a stirring, cold suspension of [PIdppe][Br] (1.85 g, 3.6 mol, 1 eq.) in THF, was added (1.81 g, 3.6 mmol, 1 eq.) of [Li][tBuCp(PPh2)2] in THF. The resulting cloudy yellow suspension was allowed to warm to room temperature immediately after addition, during which time a golden yellow solution with white precipitate appeared. The THF was removed under reduced pressure and the resulting solid was collected and placed in a soxhlet apparatus on a glass frit. The solid was washed overnight with hexanes (95 °C) to remove the dppe. The resulting solid was collected and redissolved into DCM, producing a dark yellow solution with white precipitate. This mixture was filtered through Celite® to remove the LiBr and the solution was collected. The final product was collected as a yellow powder upon removal of DCM. Single crystals were grown from the slow evaporation of the product dissolved in dichloromethane (isolated yield 1.35 g, 74%). 31P{1H} NMR (CDCl3) δ (ppm): −174.7 (t, 1Jpp = 418.3 Hz), 34.7 (d, 1Jpp = 418.5 Hz). 1H NMR (CDCl3)δ (ppm): 1.3 (s, 9H, tBu), 6.4 (t, 3JHp = 7.5 Hz, 2H, Cp–H), 7.3–7.7 (m, 20H, Ar); 13C{1H} NMR (CDCl3) δ (ppm): 32.7 (s, tBu), 33.1 (s, [C with combining low line](CH3)3), 106.8, 105.6 (dd, 1Jpc = 34.7 Hz, 2Jpc = 7.31 Hz), Cp[C with combining low line](PPh2), 111.6 (m, Cp[C with combining low line]H), 128.5 (d, 2Jpc = 13.2 Hz, o-Ph), 131.0 (s, p-Ph), 132.0 (d, 3Jpc = 10.7 Hz, m-Ph), 134 (dd, 1Jpc = 75 Hz, 4Jpc = 7 Hz, i-Ph). Anal. calcd (%) for: C33H31P3: C, 76.15; H, 6.0; N, 0; found: C, 75.68; H 6.23, N, 0.04. HR-ESI-MS: calcd for [C33H31P3]+m/z = 521.1639, found: 521.1711.

Synthesis of (tBuCp(PPh2)2P)Cr(CO)5 (2)

t BuCp(PPh2)2PI (0.048 g, 0.092 mmol, 1 eq.) and Cr(CO)6 (0.061 g, 0.277 mmol, 3 eq.) were added together to a vial in THF. The reaction solution was placed in a UV reactor under (315–400 nm) and was stirred for 3 h (the progress of the reaction can be monitored by 31P NMR). Upon completion of the reaction, the THF was removed under reduced pressure and the resulting solid was collected and any excess Cr(CO)6 was removed by sublimation (1–2 h, static vacuum, 80°). Et2O was added to the remaining solid to generate suspension which was centrifuged. The Et2O solution was collected and left to slow evaporate to deposit yellow crystalline material (yield 0.055 g, 84%). 31P{1H} NMR (CDCl3) δ (ppm): −51.0 (t, 1Jpp = 376 Hz), 28.8 (d, 1Jpp = 376 Hz). 1H NMR (CDCl3) δ (ppm): 1.3 (s, 9H, tBu), 6.5 (t, 3JHP = 4.0 Hz, 2H, Cp–H), 7.3–7.7 (m, 20H, Ar); 13C{1H} NMR (CDCl3) δ (ppm): 32.5 (s, tBu), 117.6 (m, Cp[C with combining low line]H), 129 (d, 2Jpc = 15.0 Hz, o-Ph), 132.3 (s, p-Ph), 133.0 (s, m-Ph), 218 (s, Cr([C with combining low line]O)5) FT-IR (cm−1): 2055 (s, CO), 1971 (w, CO), 1940 (s, CO), 1915 (vs, CO), 1895 (vs, CO). Anal. calcd (%) for: C38H31P3O5Cr: C, 64.05; H, 4.38; N, 0; found: C, 63.40; H 4.77, N, 0.19. HR-ESI-MS: calcd for [C38H31P3O5Cr]+m/z = 711.0918, found: 711.0915.

Synthesis of (tBuCp(PPh2)2P)W(CO)5 (3)

t BuCp(PPh2)2PI (0.048 g, 0.092 mmol, 1 eq.) and W(CO)6 (0.097 g, 0.277 mmol, 3 eq.) were added together to a vial in THF. The reaction solution was placed in a UV reactor (315–400 nm) and was stirred for 3 h (the progress of the reaction can be monitored by 31P NMR). Upon completion of the reaction, the THF was removed under reduced pressure and the resulting solid was collected and any excess W(CO)6 was removed by sublimation (1–2 h, static vacuum, 80°). Et2O was added to the remaining solid to generate suspension which was centrifuged. The Et2O solution was collected and left to slow evaporate to deposit yellow crystalline material (yield 0.061 g, 78%). 31P{1H} NMR (CDCl3)δ (ppm): −91.0 (t, 1Jpp = 356 Hz), 25.9 (d, 1Jpp = 356 Hz). 1H NMR (CDCl3)δ (ppm): 1.3 (s, 9H, tBu), 6.5 (t, 3JHP = 4.0 Hz, 2H, Cp–H), 7.5–7.8 (m, 20H, Ar); 13C{1H} NMR (CDCl3) δ (ppm): 32.5 (s, tBu), 33.1 (s, [C with combining low line](CH3)3), 111.6 (m, Cp[C with combining low line]H), 129 (s, o-Ph), 132.3 (s, p-Ph), 133.0 (s, m-Ph), 196 (s, W([C with combining low line]O)5). FT-IR (cm−1): 2065 (s, CO), 2016 (w, CO), 1980 (s, CO), 1936 (vs, CO), 1913 (vs, CO), 1891(vs, CO). Anal. calcd (%) for: C38H31P3O5W: C, 54.05; H, 3.7; N, 0; found: C, 53.58; H 3.51, N, −0.02. HR-ESI-MS: calcd for [C38H31P3O5W]+m/z = 845.0972, found: 845.0993.

Synthesis of (tBuCp(PPh2)2P)Mo2(CO)8 (4)

t BuCp(PPh2)2PI (0.048 g, 0.092 mmol, 1 eq.) and Mo(CO)6 (0.097 g, 0.277 mmol, 3 eq.) were added together to a vial in THF. The reaction solution was placed in a UV reactor (315–400 nm) and was stirred for 3 h (the progress of the reaction can be monitored by 31P NMR). Upon completion of the reaction, the THF was removed under reduced pressure and the resulting solid was collected and any excess Mo(CO)6 was removed by sublimation (1–2 h, static vacuum, 80°). Et2O was added to the remaining solid to generate suspension which was centrifuged. The Et2O solution was collected and left to slow evaporate to deposit yellow crystalline material (yield 0.050 g, 63%). 31P{1H} NMR (CDCl3)δ (ppm): −76.8 (t, 1Jpp = 366 Hz), 28.2 (d, 1Jpp = 360 Hz). 1H NMR (CDCl3)δ (ppm): 1.3 (s, 9H, tBu), 6.5 (t, 3JHP = 4.0 Hz, 2H, Cp–H), 7.5–7.6 (m, 20H, Ar); 13C{1H} NMR (CDCl3)δ (ppm): 32.4 (s, tBu), 33.1 (s, [C with combining low line](CH3)3), 103.7 (m, Cp[C with combining low line](PPh2)), 113 (s, Cp[C with combining low line]H) 128.9 (s, o-Ph), 130.8 (s, p-Ph), 132.8 (s, m-Ph), 205 (s, Mo([C with combining low line]O)5), 211 (br, Mo([C with combining low line]O)3). FT-IR (cm−1): 2067 (s, CO), 2020 (w, CO), 1981 (w, CO), 1917 (vs, CO), 1896 (vs, CO), 1831 (s, CO), 1814 (s, CO), 1804 (s, CO). Anal. calcd (%) for: C41H24P3Mo2O8: C, 52.58; H, 3.34; N, 0; found: C, 52.73; H 3.75, N, 0.23. HR-ESI-MS: calcd for [C41H24P3Mo2O8]+m/z = 937.9418, found: 937.9419.

Synthesis of (tBuCp(PPh2)2P)Mo(CO)5 (5)

t BuCp(PPh2)2PI (0.070 g, 0.134 mmol, 1 eq.), Mo(CO)6 (0.036 g, 0.134 mmol, 1 eq.) and Me3NO (0.015 g, 0.2 mmol, 1.5 eq.) were added together to a vial in THF. The reaction was left to stir for 1 h during which time bubbles could be seen forming with the evolution of CO2. The solvent was removed from the resulting gold solution, and Et2O was added to extract the product. Slow evaporation of the Et2O produced gold coloured single crystals of this material. (yield 0.082 g, 81%). 31P{1H} NMR (CDCl3)δ (ppm): −76.3 (t, 1Jpp = 355 Hz), 28.8 (d, 1Jpp = 356 Hz). 1H NMR (CDCl3)δ (ppm): 1.3 (s, 9H, tBu), 6.5 (t, 3JHP = 4.0 Hz, 2H, Cp–H), 7.5–7.7 (m, 20H, Ar); 13C{1H} NMR (CDCl3)δ (ppm): 32.5 (s, tBu), 33.0 (s, [C with combining low line](CH3)3), 103.7 (m, Cp[C with combining low line](PPh2)), 113 (m, Cp[C with combining low line]H), 128.9 (s, o-Ph), 129.0 (s, p-Ph), 132.8 (s, m-Ph), 205 (s, Mo([C with combining low line]O)5). FT-IR (cm−1): 2067 (s, CO), 2019 (w, CO), 1977 (w, CO), 1948 (s, CO), 1922 (vs, CO), 1895 (vs, CO), Anal. calcd (%) for: C38H31P3MoO5: C, 60.33; H, 4.13; N, 0; found: C, 60.7; H 4.42, N, 0.44. HR-ESI-MS: calcd for [C38H31P3MoO5]+m/z = 759.052, found: 759.0556.

Synthesis of (tBuCp(PPh2)2P)Mn2(CO)8 (6)

t BuCp(PPh2)2PI (0.053 g, 0.1 mmol, 1 eq.) and Mn2(CO)10 (0.039 g, 0.1 mmol, 1 eq.) were added together to a vial in THF to generate a pale-yellow solution. The reaction mixture was left to stir overnight, but no reaction was observed by 31P NMR. The solution was then placed in a UV reactor (315–400 nm) and stirred for 6 h to produce a red solution. The THF was removed under reduced pressure and Et2O was added to the remaining solid to generate suspension which was centrifuged. The Et2O solution was collected and left to slow evaporate, which produced large red single crystals (yield 0.058 g, 67%). 31P{1H} NMR (CDCl3)δ (ppm): 183.7 (t, 1Jpp = 283 Hz), 19 (d, 1Jpp = 283 Hz). 1H NMR (CDCl3)δ (ppm): 1.28 (s, 9H, tBu), 6.5 (t, 3JHP = 4.2 Hz, 2H, Cp–H), 7.3–7.8 (m, 20H, Ar); 13C{1H} NMR (CDCl3)δ (ppm): 29.8 (s, tBu), 32.4 (s, [C with combining low line](CH3)3), 117 (s, Cp[C with combining low line]H), 129 (m, o-Ph), 133 (s, p-Ph), 134.2 (s, m-Ph), 219 (br, Mn([C with combining low line]O)). FT-IR (cm−1): 2050 (s, CO), 1988 (s, CO), 1946 (vs, CO), 1930 (vs, CO), 1911 (vs, CO). Anal. calcd (%) for: C41H24P3Mn2O8: C, 57.63; H, 3.66; N, 0; found: C, 57.1; H 3.71, N, 0.26. HR-ESI-MS: calcd for [C41H24P3Mn2O8]+m/z = 855.0069, found: 855.0066.

Synthesis of (tBuCp(PPh2)2P)Fe(CO)4 (7)

t BuCp(PPh2)2PI (0.052 g, 0.099 mmol, 1 eq.) and Fe2(CO)9 (0.072 g, 0.199 mmol, 2 eq.) were added together to a vial in THF. The reaction was left to stir for 4 days. The resulting solution was centrifuged and the THF was removed in vacuo along with any Fe(CO)5 that may have formed because of the excess Fe2(CO)9 which was used in the reaction. Hexanes was then added to the precipitate to extract the product. A small impurity remains in the extracted solution, and has the same solubility as the product (both are soluble in both polar and non-polar solvents), thus far we have been unsuccessful in the separation of the bulk material. Single crystalline material of 8 was obtained from the slow evaporation of the product in hexanes. 31P{1H} NMR (CD2Cl2)δ (ppm): −16 (t, 1Jpp = 374 Hz), 22 (d, 1Jpp = 374 Hz). 1H NMR (CD2Cl2)δ (ppm): 1.30 (s, 9H, tBu), 6.4 (t, 3JHP = 5 Hz, 2H, Cp–H), 7.4–7.7 (m, 20H, Ar); 13C{1H} NMR (CD2Cl2)δ (ppm): 29 (s, tBu), 30 (s, [C with combining low line](CH3)3), 114 (s, Cp[C with combining low line]H), 128 (m, o-Ph), 132 (m, p-Ph), 133 (m, m-Ph), 214 (br, Fe([C with combining low line]O)). FT-IR (cm−1): 2033 (s, C–O), 2007 (s, C–O), 1981 (s, C–O), 1959 (vs, C–O), 1930 (vs, C–O).

Synthesis of (tBuCp(PPh2)2P)Co2(CO)6 (8)

t BuCp(PPh2)2PI (0.018 g, 0.0345 mmol, 1 eq.) and Co2(CO)8 (0.014 g, 0.0345 mmol, 1.1 eq.) were added together to a vial in MeCN, and the solution immediately turns dark red in colour and bubbles are formed, signifying the loss of CO. The proligand was completely consumed after 5 min, as confirmed by 31P NMR. The solution was then centrifuged to remove black precipitate that had formed and the MeCN was left to slowly evaporate to generate red single crystals. Unfortunately, we have been unable to obtain elemental analysis or mass spectroscopy on this material; we postulate that the complex eventually decomplexes, which has inhibited our ability to obtain microanalysis (yield 0.015 g, 56%). 31P{1H} NMR (CD3CN)δ (ppm): 177.0 (t, 1Jpp = 242 Hz), 6.0 (d, 1Jpp = 235 Hz). 1H NMR (CD3CN)δ (ppm): 1.4 (s, 9H, tBu), 6.9 (t, 1JHP = 4 Hz, 2H, Cp-H), 7.4–7.4 (m, 20H, Ar); 13C{1H} NMR (CD2Cl2)δ (ppm): 32.0 (s, tBu), 32.9 (s, [C with combining low line](CH3)3), 117 (t, 2JPC = 8.75 Hz, Cp[C with combining low line]H), 126 (m, Cp[C with combining low line](PPh2)), 128 (m, o-Ph), 129 (s, p-Ph), 133 (s, m-Ph), 205 (br, Co([C with combining low line]O)). FT-IR (cm−1): 2037 (s, CO), 1998 (vs, CO), 1979 (s, CO), 1962 (vs, CO), 1932 (s, CO).

Conflicts of interest

There are no conflicts to declare.

References

  1. M. Shibasaki, M. Kanai, S. Matsunaga and N. Kumagai, Acc. Chem. Res., 2009, 42, 1117–1127 CrossRef CAS PubMed .
  2. P. Buchwalter, J. Rose and P. Braunstein, Chem. Rev., 2015, 115, 28–126 CrossRef CAS PubMed .
  3. P. J. Steel, Acc. Chem. Res., 2005, 38, 243–250 CrossRef CAS PubMed .
  4. B. Pan, D. A. Evers-McGregor, M. W. Bezpalko, B. M. Foxman and C. M. Thomas, Inorg. Chem., 2013, 52, 9583–9589 CrossRef CAS PubMed .
  5. D. Morales-Morales and C. G. M. Jensen, The Chemistry of Pincer Compounds, Elsevier, 2011 Search PubMed .
  6. B. K. Langlotz, H. Wadepohl and L. H. Gade, Angew. Chem., Int. Ed., 2008, 47, 4670–4674 CrossRef CAS PubMed .
  7. G. Van Koten, in Organometallic pincer chemistry, Springer, 2013, pp. 1–20 Search PubMed .
  8. E. M. Schuster, M. Botoshansky and M. Gandelman, Angew. Chem., Int. Ed., 2008, 47, 4555–4558 CrossRef CAS PubMed .
  9. J. Park and S. Hong, Chem. Soc. Rev., 2012, 41, 6931–6943 RSC .
  10. R. A. Jones, A. L. Stuart, J. L. Atwood, W. E. Hunter and R. D. Rogers, Organometallics, 1982, 1, 1721–1723 CrossRef CAS .
  11. C. Mealli, A. Ienco, A. Galindo and E. P. Carreño, Inorg. Chem., 1999, 38, 4620–4625 CrossRef CAS PubMed .
  12. R. G. Hayter, J. Am. Chem. Soc., 1963, 85, 3120–3124 CrossRef CAS .
  13. R. G. Hayter, Inorg. Chem., 1963, 2, 1031–1035 CrossRef CAS .
  14. C. Kling, D. Leusser, T. Stey and D. Stalke, Organometallics, 2011, 30, 2461–2463 CrossRef CAS .
  15. F. Mathey, J. Organomet. Chem., 1994, 475, 25–30 CrossRef CAS .
  16. F. Mathey, Coord. Chem. Rev., 1994, 137, 1–52 CrossRef CAS .
  17. C. Elschenbroich, M. Nowotny, B. Metz, W. Massa, J. Graulich, K. Biehler and W. Sauer, Angew. Chem., Int. Ed. Engl., 1991, 30, 547–550 CrossRef .
  18. F. Nief, C. Charrier, F. Mathey and M. Simalty, J. Organomet. Chem., 1980, 187, 277–285 CrossRef CAS .
  19. K. Dimroth and H. Kaletsch, J. Organomet. Chem., 1983, 247, 271–285 CrossRef CAS .
  20. K. C. Nainan and C. T. Sears, J. Organomet. Chem., 1978, 148, C31–C34 CrossRef CAS .
  21. J. Deberitz and H. Nöth, J. Organomet. Chem., 1973, 49, 453–468 CrossRef CAS .
  22. J. Deberitz and H. Nöth, Chem. Ber., 1970, 103, 2541–2547 CrossRef CAS .
  23. A. J. Arce, A. J. Deeming, Y. De Sanctis and J. Manzur, J. Chem. Soc., Chem. Commun., 1993, 5, 325 RSC .
  24. C. Müller and D. Vogt, Dalton Trans., 2007, 9226, 5505 RSC .
  25. L. Boubekeur, L. Ricard, P. Le Floch and N. Mézailles, Organometallics, 2005, 24, 3856–3863 CrossRef CAS .
  26. J. Grobe, D. Le Van, J. Nientiedt, B. Krebs and M. Dartmann, Chem. Ber., 1988, 121, 655–664 CrossRef CAS .
  27. L. Weber, Eur. J. Inorg. Chem., 2000, 2000, 2425–2441 CrossRef .
  28. L. Weber, J. Krümberg, H.-G. Stammler and B. Neumann, Z. Anorg. Allg. Chem., 2004, 630, 2478–2482 CrossRef CAS .
  29. L. Weber, U. Lassahn, H.-G. Stammler, B. Neumann and K. Karaghiosoff, Eur. J. Inorg. Chem., 2002, 2002, 3272–3277 CrossRef .
  30. D. Gudat, M. Schrott, V. Bajorat and M. Nieger, Phosphorus, Sulfur Silicon Relat. Elem., 1996, 109, 125–128 CrossRef .
  31. D. Gudat, M. Schrott and M. Nieger, J. Chem. Soc., Chem. Commun., 1995, 371, 1541–1542 RSC .
  32. D. Gudat, M. Schrott, V. Bajorat, M. Nieger, S. Kotila, R. Fleischer and D. Stalke, Chem. Ber., 1996, 129, 337–345 CrossRef CAS .
  33. D. Gudat, M. Nieger and M. Schrott, Inorg. Chem., 1997, 36, 1476–1481 CrossRef CAS PubMed .
  34. J. D. Protasiewicz, Eur. J. Inorg. Chem., 2012, 29, 4539–4549 CrossRef .
  35. L. Liu, D. A. Ruiz, F. Dahcheh and G. Bertrand, Chem. Commun., 2015, 51, 12732–12735 RSC .
  36. M. Alcarazo, K. Radkowski, G. Mehler, R. Goddard and A. Fürstner, Chem. Commun., 2013, 49, 3140–3142 RSC .
  37. D. V. Partyka, M. P. Washington, J. B. Updegraff, R. A. Woloszynek and J. D. Protasiewicz, Angew. Chem., Int. Ed., 2008, 47, 7489–7492 CrossRef CAS PubMed .
  38. B. A. Surgenor, B. A. Chalmers, K. S. Athukorala Arachchige, A. M. Z. Slawin, J. D. Woollins, M. Bühl and P. Kilian, Inorg. Chem., 2014, 53, 6856–6866 CrossRef CAS PubMed .
  39. A. Doddi, D. Bockfeld, A. Nasr, T. Bannenberg, P. G. Jones and M. Tamm, Chem. – Eur. J., 2015, 21, 16178–16189 CrossRef CAS PubMed .
  40. V. A. K. Adiraju, M. Yousufuddin and H. V. Rasika Dias, Dalton Trans., 2015, 44, 4449–4454 RSC .
  41. O. Lemp and C. von Hänisch, Phosphorus, Sulfur, Silicon Relat. Elem., 2016, 191, 659–661 CrossRef CAS .
  42. F. Mathey, Angew. Chem., Int. Ed. Engl., 1987, 26, 275–286 CrossRef .
  43. F. Mathey, Dalton Trans., 2007, 2007, 1861–1868 RSC .
  44. D. V. Partyka, M. P. Washington, J. B. U. Iii, R. A. Woloszynek and J. D. Protasiewicz, Angew. Chem., Int. Ed., 2008, 47, 7489–7492 CrossRef CAS PubMed .
  45. W. Petz and G. Frenking, Top. Organomet. Chem., 2010, 49–92 CrossRef CAS .
  46. M. Alcarazo, K. Radkowski, G. Mehler, R. Goddard and A. Fürstner, Chem. Commun., 2013, 49, 3140–3142 RSC .
  47. P. K. Coffer (née Monks) and K. B. Dillon, Coord. Chem. Rev., 2013, 257, 910–923 CrossRef .
  48. J. W. Dube, C. L. B. Macdonald and P. J. Ragogna, Angew. Chem., Int. Ed., 2012, 51, 13026–13030 CrossRef CAS PubMed .
  49. P. K. Coffer née Monks, R. M. K. Deng, K. B. Dillon, M. A. Fox and R. J. Olivey, Inorg. Chem., 2012, 51, 9799–9808 CrossRef PubMed .
  50. J. W. Dube, M. M. Hänninen, J. L. Dutton, H. M. Tuononen and P. J. Ragogna, Inorg. Chem., 2012, 51, 8897–8903 CrossRef CAS PubMed .
  51. S. C. Kosnik, G. J. Farrar, E. L. Norton, B. F. T. Cooper, B. D. Ellis and C. L. B. Macdonald, Inorg. Chem., 2014, 11, 13061–13069 CrossRef PubMed .
  52. S. C. Kosnik and C. L. B. Macdonald, Dalton Trans., 2016, 45, 6251–6258 RSC .
  53. R. Broussier, E. Bentabet, P. Mellet, O. Blacque, P. Boyer, M. M. Kubicki and B. Gautheron, J. Organomet. Chem., 2000, 598, 365–373 CrossRef CAS .
  54. F. H. Allen, Acta Crystallogr., Sect. B: Struct. Sci., 2002, 58, 380–388 CrossRef .
  55. H. Schumann, L. Rösch, H.-J. Kroth, H. Neumann and B. Neudert, Chem. Ber., 1975, 108, 2487–2499 CrossRef CAS .
  56. J. C. Mitchener and M. S. Wrighton, J. Am. Chem. Soc., 1983, 105, 1065–1067 CrossRef CAS .
  57. C. S. Kraihanzel and F. A. Cotton, Inorg. Chem., 1963, 2, 533–540 CrossRef CAS .
  58. H. Schumann, L. Rüsch, H.-J. Kroth, J. Pickardt, H. Neumann and B. Neudert, Z. Anorg. Allg. Chem., 1977, 430, 51–60 CrossRef CAS .
  59. M. F. Guns, E. G. Claeys and G. P. Van Der Kelen, J. Mol. Struct., 1979, 53, 45–53 CrossRef CAS .
  60. D. Bockfeld, A. Doddi, P. G. Jones and M. Tamm, Eur. J. Inorg. Chem., 2016, 2016, 3704–3713 CrossRef CAS .
  61. J. W. Dube, C. L. B. Macdonald, B. D. Ellis and P. J. Ragogna, Inorg. Chem., 2013, 52, 11438–11449 CrossRef CAS PubMed .
  62. S. C. Kosnik, J. F. Binder, M. C. Nascimento and C. L. B. Macdonald, J. Visualized Exp., 2016, 3–9 Search PubMed .
  63. D. Kratzert, J. J. Holstein and I. Krossing, J. Appl. Crystallogr., 2015, 48, 933–938 CAS .
  64. C. B. Hübschle, G. M. Sheldrick and B. Dittrich, J. Appl. Crystallogr., 2011, 44, 1281–1284 CrossRef PubMed .
  65. M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, J. A. Montgomery Jr., J. E. Peralta, F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, N. Rega, N. J. Millam, M. Klene, J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels, Ö. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski and D. J. Fox, Gaussian09, Revision D.01, Gaussian Inc., Wallingford, CT, 2009 Search PubMed .
  66. J. Perdew, K. Burke and M. Ernzerhof, Phys. Rev. Lett., 1996, 77, 3865–3868 CrossRef CAS PubMed .
  67. J. P. Perdew, K. Burke and M. Ernzerhof, Phys. Rev. Lett., 1997, 1396 CrossRef CAS .
  68. C. Adamo and V. Barone, J. Chem. Phys., 1999, 110, 6158 CrossRef CAS .
  69. A. Schafer, H. Horn and R. Ahlrichs, J. Chem. Phys., 1992, 97, 2571–2577 CrossRef .
  70. A. Schäfer, C. Huber and R. Ahlrichs, J. Chem. Phys., 1994, 100, 5829–5835 CrossRef .
  71. M. Dolg, H. Stoll, A. Savin and H. Preuss, Theor. Chim. Acta, 1989, 75, 173–194 CrossRef CAS .
  72. M. Dolg, U. Wedig, H. Stoll and H. Preuss, J. Chem. Phys., 1987, 86, 866–872 CrossRef CAS .
  73. Gaussview 3.0, Gaussian Inc., Pittsburgh, PA, 2003 Search PubMed .
  74. A. E. Reed, L. A. Curtiss and F. Weinhold, Chem. Rev., 1988, 88, 899–926 CrossRef CAS .
  75. E. D. Glendening, A. E. Reed, J. E. Carpenter and F. Weinhold, NBO 3.0 Search PubMed .
  76. W. Humphrey, A. Dalke and K. Schulten, J. Mol. Graphics, 1996, 14, 33–38 CrossRef CAS PubMed .

Footnote

Electronic supplementary information (ESI) available. CCDC 1578538 and 1578540–1578546. For ESI and crystallographic data in CIF or other electronic format see DOI: 10.1039/c7dt03844e

This journal is © The Royal Society of Chemistry 2017